If you've given up on Google parent company Alphabet’s Project Loon ever seeing the light of day, you might want to rethink that position. The project, which is part of Google's experimental "X" division, announced yesterday that it has had a breakthrough that could allow the technology to come to market sooner than expected.

Project Loon (pictured above) is Google's plan to provide low-cost Internet access to most of the planet by launching a fleet of Internet-enabled access points in hot air balloons that would beam connectivity down to the surface from their positions in the stratosphere. The ambitious proposal faces a number of engineering challenges, including the difficulty in maintaining balloons designed to stay aloft for months at a time.

Clusters of Balloons

But the project may have been able to overcome some of the most critical challenges it faced thanks to some recent advancements in artificial intelligence and machine learning, according to a blog post by Astro Teller, who has been in charge of Google X labs since 2010. Teller described the development as an "unexpected discovery" that is propelling the project forward faster than the team at X had anticipated.

The breakthrough consists of a change to the project's navigational algorithm, which is now capable of sending small teams of balloons to form a cluster of access points over a specific geographic region, according to Teller. "This is a shift from our original model for Loon in which we planned to create rings of balloons sailing around the globe, and balloons would take turns moving through a region to provide service," Teller said.

As a result, Teller said the company should be able to deploy a viable network to a target region in a matter of weeks, rather than months that would involve fewer balloons, offer easier management of the operation and cost less. Teller said that was crucial, as cost has been one of the most critical factors keeping Project Loon from becoming reality.

Greater Precision

The original idea was for a much larger fleet of balloons to float freely in the stratosphere, acting as flying cell phone towers. Google X had hoped to be able to control the balloons just enough so that as one floated out of range of a particular region, another would come into range.

Instead, the team has found that they are able to control the balloons much more precisely, allowing them to direct coverage to specific regions to within tens of kilometers of their target.

The team then discovered how to keep the balloons clustered together in a particular area as a single group for up to three months by having them fly around in increasingly smaller loops. One of its balloons was even able to maintain a position above Peruvian airspace for 98 days.

The breakthrough means that Loon will be able to run experiments in providing Internet with as few as 30 balloons, rather than as many as 400 as the team originally believed. That could mean cheap, high-speed Internet access might be on its way a lot faster than anyone ever imagined.